💡 Welcome to “Help me Build my Workflow”

[bartv]

Need help creating an n8n workflow? Stuck on a tricky automation? This is the place to collaborate with other community members and build your workflows in public!

Whether you’re refining a complex automation or just getting started, our community is here to brainstorm, troubleshoot, and share insights.

How to Get the Best Help

To make sure others can support you effectively, please:

Describe your goal clearly – What problem are you trying to solve?
Show what you’ve tried – Share your current setup, what works, and where you’re stuck.
Include details – Screenshots, error messages, or JSON exports can help others understand your issue faster.

Rules for this Category

No paid work requests or offers – This is a community space for knowledge-sharing, not a job board.
No direct requests to n8n Staff – We love our community, but this category is for peer-to-peer help. Please don’t tag n8n team members or send direct messages requesting support.

Let’s build amazing workflows together!

[Block_Rocks]

Trying to do this since 1 week, no previously exp here, few erros but Im exhausted and tried hard last 2 days, please help me tomorrow cause ill sleep in 3 2 zzzzzzzzzzzzzzzzzzzzz

Automate Multi-Platform Social Media Content Creation with AI

[n8nstein]

What’s the issue? Can you share the json to look into this further?

[bryanham1]

My problem is that I am trying to call a workflow from a visual studio webform using vb.net code. I tried using both process.start(workflowurl) and a hyperlink using the workflow url. Both of those approches are a “Get” call, while my workflow is to allow downloading a file using an ftp node and a response node, which needs to be a “Post” request. Does not work. I also tried the following code using httpclient and it does not work. Do you have a way to call an ftp workflow from a webform using vb. net code? Private Sub Button6_Click(sender As Object, e As EventArgs) Handles Button6.Click
Page.RegisterAsyncTask(New PageAsyncTask(AddressOf MyAsyncMethod))
End Sub

Private Async Function MyAsyncMethod() As Task

'Dim webhookUrl As String = "http://localhost:5678/webhook-test/823f20b4-f8fd-4736-b257-d9418dccffd8" ' Replace with your actual webhook URL
Dim webhookUrl As String = "https://silly-eel-99.hooks.n8n.cloud/webhook/fbcd3019-1603-4a1c-8f41-941bfb55387e"

'Dim jsonPayload As String = "{""name"": ""John Doe"", ""email"": ""[email protected]""}" ' Example JSON data to send

Using httpClient As New HttpClient()
    Try
        'Dim content As New StringContent(jsonPayload, Encoding.UTF8, "application/json") ' Set content type to JSON

        'Dim response As HttpResponseMessage = Await httpClient.PostAsync(webhookUrl, Content) ' Make the POST request
        Dim response As HttpResponseMessage = Await httpClient.GetAsync(webhookUrl) ' Make the POST request

        If response.IsSuccessStatusCode Then
            ' Workflow triggered successfully
            Console.WriteLine("n8n workflow triggered successfully!")
        Else
            ' Handle potential errors (e.g., incorrect URL, network issues)
            Console.WriteLine($"Error triggering n8n workflow: {response.StatusCode} - {Await response.Content.ReadAsStringAsync()}")
        End If

    Catch ex As Exception
        ' Handle any exceptions during the HTTP request
        Console.WriteLine($"An error occurred: {ex.Message}")
    End Try
End Using
Await Task.Delay(10000) ' Example delay

End Function

[CoolMcCool99]

Hello, can you help me build a completely local workflow to analyze IPs, URLs, files, etc., and see if they are malicious? I know there is already a workflow, but it is not completely local.

Let me explain: I have an environment in which I have a server with Docker and n8n locally, as well as AI, but the AI is on another completely isolated server and the only way to communicate is through port 5678.

With that in mind, what could the workflow be?

[Some_One]

i tried to ask a question but it gave me this Token type `html_raw` not supported by Markdown parser so i cant even ask my question about my work flow

[Chris_Brown]

Hi I have a gmail Imap Trigger that was working fine, but today i started getting this error: {
“errorMessage”: “Search option argument must be a Date object or a parseable date string”,
“errorDetails”: {},
“n8nDetails”: {
“n8nVersion”: “1.113.3 (Self Hosted)”,
“binaryDataMode”: “default”,
“stackTrace”: [
“Error: Search option argument must be a Date object or a parseable date string”,
" at buildSearchQuery (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/[email protected]/node_modules/imap/lib/Connection.js:1932:21)“,
" at Connection.search (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/[email protected]/node_modules/imap/lib/Connection.js:577:15)“,
" at Connection.search (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/[email protected]/node_modules/imap/lib/Connection.js:566:8)”,
" at /usr/local/lib/node_modules/n8n/node_modules/.pnpm/@n8n+imap@file+packages+@n8n+imap/node_modules/@n8n/imap/src/imap-simple.ts:73:14",
" at new Promise ()“,
" at ImapSimple.search (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/@n8n+imap@file+packages+@n8n+imap/node_modules/@n8n/imap/src/imap-simple.ts:72:16)”,
" at TriggerContext.getNewEmails (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/n8n-nodes-base@file+packages+nodes-base@[email protected]_asn1.js@5_afd197edb2c1f848eae21a96a97fab23/node_modules/n8n-nodes-base/nodes/EmailReadImap/v2/utils.ts:114:34)”,
" at Connection.onMail (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/n8n-nodes-base@file+packages+nodes-base_@[email protected]_asn1.js@5_afd197edb2c1f848eae21a96a97fab23/node_modules/n8n-nodes-base/nodes/EmailReadImap/v2/EmailReadImapV2.node.ts:422:27)“,
" at Connection.emit (node:events:531:35)”,
" at Connection._resUntagged (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/[email protected]/node_modules/imap/lib/Connection.js:1287:14)"
]
}
}

There are no dates being searched for

Im hoping someone can assist

Regards

John

[Globalistik]

Hi community ,

I’m building a workflow in n8n to centralize transportation requests. The current structure is:
Gmail Trigger → Basic LLM Chain (with Structured Output Parser.1) → Edit Fields → Google Sheets

The issue I’m facing is that I cannot connect the Structured Output Parser.1 directly to the Edit Fields node. n8n only allows me to connect Basic LLM Chain → Edit Fields, but when I do this, I get the following error:
No data found from “main” input

I understand this means the Edit Fields node is not receiving any valid input items, but I don’t understand how to make the structured JSON from Structured Output Parser.1 available as a main output to use it in Edit Fields or in a Set node.

In the output of Structured Output Parser.1 I can clearly see my structured JSON inside response.output, for example:

In the Edit Fields node, I’m trying to use expressions like:

But since Edit Fields is only connected to the Basic LLM Chain, I either get undefined or the “main input” error

Question:

How should I correctly connect the Structured Output Parser.1 so that its data is available in the Edit Fields or Set node?
Is there a way for the parser to expose a main output that can be connected, or should I always map directly from the Basic LLM Chain output?

Any guidance would be super helpful

[skullsaints151413]

sure why not

[Clair_Balkowski]

Clair_Balkowski

1

20h

I am having the same issue and not seeing any icons or error tryangles beside the workflow … i am new to this n8n and dont have a great understanding but would like further help in this as i am trying to utilize the Auto Social Media Posts AI Agent and i have everything connected…
Please help

these are the steps I have followed -https://www.youtube.com/watch?v=r2bxV2CHu5U

now i have tried to exacute workflow and it shows this error “ Problem running workflow

Please resolve outstanding issues before you activate it”

Thanks

HERE IS MY WORKFLOW :

[hjam10]

Help: Setting up Mistral OCR workflow to extract structured data from complex image documents (e.g., vehicle registrations)

Hi everyone

I’m currently building a workflow to automatically extract text and structured fields from complex image documents, specifically vehicle registration files.

The goal is to use the Mistral AI “Extract Text” node (or Mistral OCR API via HTTP Request) to read the document and output clean, structured JSON that can be mapped into an Excel or database table.

The outputs are not consistent and not detailed at all. Is there any way reliable and precise text extraction of such image files?

Thank You.

Workflow:

example file:

Goal output: JSON like

{
  "plateNumber": "LI-D715",
  "vin": "W1V44770513900923",
  "firstRegistration": "13.12.2022",
  "vehicleModel": "Mercedes-Benz Vito Tourer",
  "ownerName": "HOCH- UND INGENIEURBAU GMBH",
  "ownerAddress": "Abt-Knauer-Str. 50, 96260 Weismain"
}

Current output:

“output”: {

“Name”: “”,

“Erstzulassung”: “12.12.2022”,

“Adresse”: “”,

“Fahrzeugidentifikationsnummer”: “AAQ00882”,

“Fahrzeugname”: “VITO TOURER”,

“Marke”: “MERCEDES-BENZ”,

“Motorcode”: “GD339001”

[Florian_Moritz]

Need advice on n8n data architecture for AI Agent with Supabase Edge Functions (booking/calendar use case)
We’re building an AI-powered booking assistant in n8n that connects to Supabase Edge Functions for all business logic - like reading and writing to a booking calendar system (similar to what a nail studio would use).

The agent runs through a chat interface (currently via the Chat Trigger, later WhatsApp) and can book, update, or delete appointments in Supabase.
Each action requires structured data such as client_id, service_id, appointment_date, and team_member_id, which the Edge Functions validate and handle (business hours, conflicts, etc.).

Our current challenge is data persistence and memory:

• The Booking Agent can reason and call the correct Edge Functions, but loses context between messages (e.g. forgets the client_id or name, even though it’s stored in Supabase).

• We already write to a conversation state table and read from it on each turn, but the merge logic resets some fields.

We’d love community input on the best n8n data architecture to:

1. Persist structured context (IDs, slots) between chat turns reliably

2. Maintain short-term conversational memory (last few exchanges)

3. Keep the flow efficient using one main AI agent (not multiple sub-agents)

Has anyone implemented a similar architecture - especially combining Postgres Chat Memory or session-based context retrieval with Edge Functions for logic execution?
Any suggestions or best-practice references for designing this kind of memory architecture would be highly appreciated.

Attached screenshot: current n8n workflow setup - Chat Trigger, Booking Agent, Edge Functions, and Memory nodes.)

[Bharat_Varshney]

I have use case and I am sharing my json here is the link https://drive.google.com/file/d/1hVueMu4kNQIEUcBvywaknrL5LiYp6TWV/view?usp=sharing and use case is this -A requester fills out a form with details about what they need to buy (item, cost, supplier, etc.).

The workflow checks if approval is needed. If the amount is high or certain rules apply, it sends a request to a manager. Low-cost requests can be auto-approved.

Once approved, the system automatically generates a purchase order (PO) number and records the order details.

When goods are delivered, the receiver marks them as ‘received’ in the system, confirming the PO.

A supplier can submit an invoice using another simple form, which is linked to the PO and the delivery info.

The workflow checks all three: PO, receipt, and invoice—this is called “three-way matching.” If everything matches (amount, items, supplier),

The invoice is approved and marked for payment. If something doesn’t match, the system flags it for review

[Bharat_Varshney]

@Clair_Balkowski @Florian_Moritz @bartv @Chris_Brown Can you please help

[manav]

plz help me in fixing it